Hydrogels Poly(vinyl alcohol)/Gentamicin and Poly(vinyl alcohol)/Chitosan/Gentamicin: A Promising Approach to Accelerate Burn Wound Healing

1. Introduction

Scar formation and delayed wound healing are significant challenges in treating skin tissue trauma [1,2]. Wound healing and tissue regeneration are a dynamic physiological process influenced by the interactions of the extracellular matrix, various cell types, and growth factors [3]. In severe conditions like trauma, diabetic or burn wounds, the normal processes of re-epithelialization and dermal repair may be insufficient for healing ([4]. Debridement, pressure reduction, and infection prevention are traditional wound treatment methods [5]. While effective for most injuries, larger skin wounds may need dressings that mimic skin tissue to support cell adhesion, migration, proliferation, and tissue generation by mimicking the extracellular matrix (ECM) [6]. Recent advancements in biomimetic scaffolds utilizing stem cells and bioactive substrates have created promising therapeutic options for treating full-thickness skin injuries [7]. Various materials and structures, such as hydrogels, films, and foams, are used as scaffolds to enhance healing and protect the skin from additional damage, because they possess ECM-like structure, high porosity, and permeability [6]. Hydrogels are regarded for ideal skin substitutes due to their moisture retention, effective fluid absorbance, and high water retention capacity [8,9]. Their porous structure provides support in absorbing wound exudate, minimizing infection risk, and promoting an environment conducive to wound healing [8,9]. Additionally, utilization of natural-synthetic polymer composites in hydrogel fabrication results in a porous structure that benefit from the biocompatibility of natural polymers and the customizable properties of synthetic polymers [10].

Poly(vinyl alcohol) (PVA) is a synthetic polymer widely used in tissue engineering due to its excellent biocompatibility, biodegradability, non-toxicity, non-carcinogenicity, and water solubility [11]. This cost-effective polymer possesses favorable physical properties, including good transparency, low interfacial tension, and a high swelling ratio. To improve mechanical properties, PVA has often been combined with natural polymers, as is PVA/chitosan composite hydrogel, with enhances softness and flexibility [10,12].

Chitosan (CHI), a polysaccharide-based polymer, is particularly effective in wound healing due to its ability to reduce pain by blocking nerve endings [12]. It accelerates natural blood clot formation and minimizes scarring [13]. Chitosan-based hydrogels possess excellent bioadhesive properties, prevent microbial penetration, control inflammation, and enhance natural hyaluronic acid levels at the wound site [14]. Structurally similar to glycosaminoglycans in the ECM, CHI boosts fibroblast activation, regulates collagen fiber density and distribution, and promotes cell migration, granulation tissue formation, and vascularization - essential processes in wound healing [13]. Its stimulating effect on leukocytes and potent antibacterial properties make CHI a widely used biomaterial in wound care [14].

Gentamicin (Gent) is an aminoglycoside antibiotic known for its effective antimicrobial activity, making it widely used in treating microbial infections, particularly burn wounds [15]. However, its ability to penetrate the deeper layers of skin is limited due to low systemic absorption, likely because of its cationic nature [16,17]. Consequently, Gent primarily exerts its effects on the superficial skin layer. Despite its efficacy, Gent can cause renal tubular necrosis and congestion [18], as well as harm intra-auricular lymphocytes, leading to nephrotoxicity and ototoxicity, which somewhat restricts its clinical use [19]. To mitigate these toxic effects, researchers have explored methods to control Gent release by embedding or fixing it, allowing for sustained action [20]. Alternatively, Gent can be delivered directly to the target site for specific binding, enhancing its prolonged effect [21].

The aim of this study was to synthesize PVA/Gent and PVA/CHI/Gent hydrogels, and conduct an in vivo experiment using a rat model to investigate the potential promoting effect of the PVA/Gent and PVA/CHI/Gent hydrogels on burn wound healing. These innovative hydrogels, originally synthesis in our laboratory and applied for the first time in in vivo experiments, should enable the long-term controlled release of embedded Gent in the wound locally, avoiding the systemic antibiotic administration.

2. Results and Discussion

2.1. Antibacterial Activity

Prior to the conducting of the in vivo experiment antibacterial activity of synthesized hydrogels with different combinations of substances (PVA, CHI and Gent) were examined.

Figure 1 represents photographs of disc-diffusion tests against E. coli and S. aureus of PVA, PVA/Gent, PVA/CHI and PVA/CHI/Gent hydrogels.

Hydrogel comprised of PVA did not exhibit any antibacterial effect on the tested bacterial strains. Evidently, all the samples with CHI showed certain antibacterial activity, which was stronger against E. coli than S. aureus. The average width of the inhibition zone against S. aureus was 12 mm. In the case of E. coli, two distinct regions of sensitivity are observed; highly sensitive zone of 12 mm and the moderate sensitivity zone of 14 mm in diameter. As expected, the zones of inhibition are clearly visible for all Gent containing hydrogels. Hydrogel PVA/Gent had a pronounced highly sensitive zone of 24.5 mm and the moderate sensitivity zone of 34 mm in diameter when tested against E. coli. The same sample acted very efficiently against S. aureus, since the inhibition zone was 31.5 mm in diameter. In the case of PVA/CHI/Gent, the addition of antibiotic increased the antibacterial activity of hydrogels. Against S. aureus PVA/CHI/Gent gave 31 mm wide zone of inhibition. When tested against E. coli difference in bacterial sensitivity was observed with a lighter colored, highly sensitive zone of 31 mm and the darker colored, moderate sensitivity zone of 33 mm in diameter (Figure 1 A, B).

The results obtained by disc-diffusion test are in accordance with the kinetics of antibacterial activity against E. coli ATCC25922 and S. aureus [22]. In the case of Gram-negative E. coli, PVA/Gent and PVA/CHI/Gent hydrogels exhibited a bactericidal effect, since the reduction of the number of viable bacterial colonies was more than three orders of magnitude after only 15 min of incubation. After 1h of inoculation, a sterile environment was achieved since there were no longer any live E. coli cells present. For the Gram-positive S. aureus, the absence of any live cells was observed after only 15 min. The mechanisms of antibacterial activity can be explained as follows. Gentamicin exerts its antibacterial effect by irreversibly binding to the 30S ribosomal subunit and 16S rRNA in bacterial cells. This interaction disrupts tRNA recognition, leading to misreading of mRNA and preventing the synthesis of essential proteins. Specifically, Gent binds to four nucleotides of 16S rRNA and an amino acid in the S12 protein, interfering with the ribosomal decoding site. As a result, incorrect amino acids are incorporated into the growing polypeptide chain, ultimately disrupting protein function and bacterial survival [23].

2.2. Gentamicin Release

To investigate the kinetics and mechanism of Gent release from PVA/Gent and PVA/CHI/Gent hydrogels, the experimental releasing profile of Gent has been determined using HPLC techniques, as the time dependence of the ratio c_t/c_o, where c_t is the concentration of Gent released from hydrogel at time, t and c_o is the initial concentration of Gent inside the hydrogel [22]. The experimental data were compared to several theoretical models in order to elucidate the Gent diffusion coefficient. The models applied were Makoid-Banakar [24], Korsmeyer-Peppas [25] and Kopcha [26] described by equations Eqs. (1), (2) and (3), respectively:

$${\displaystyle \frac{C_{\mathrm{t}}}{C_0}}={\mathrm{k}}_{\mathrm{M}\mathrm{B}}\cdot t^{\mathrm{n}}\cdot \exp \left(-c\cdot t\right)$$

(1)

$${\displaystyle \frac{C_{\mathrm{t}}}{C_0}}=k_{\mathrm{K}\mathrm{P}}\cdot t^{\mathrm{n}}$$

(2)

$${\displaystyle \frac{C_{\mathrm{t}}}{C_0}}=\mathrm{A}\cdot t^{1/2}+\mathrm{B}\cdot t$$

(3)

where k_MB is the Makoid-Banakar constant; c is the Makoid-Banakar parameter; k_KP is the Korsmeyer-Peppas constant; n - the coefficient which describes release transport mechanism (n<0.5 - Fickian diffusion, n>0.5 - non-Fickian/anomalous diffusion, n=1 - Case II transport) [25]; A and B - Kopcha’s constants which depend on the dominant transport phenomenon during release.

Gentamicin release profiles verified the initial burst release effect of Gent from the hydrogel, i.e., 70% loaded antibiotic was released within the first 48h which could be very useful in preventing biofilm formation, followed by slow release of Gent in a later time period. The time exponent, n, is an indication of the dominant diffusion mechanism and, as its values were less than 0.5, it can be concluded that the release of Gent from hydrogel conformed to the Fickian diffusion behavior [25] and was governed mainly by the concentration gradient of released Gent. This was also proved by Kopcha model, as the absolute values of parameter A were higher compared to B, indicating that the predominant driving force for the release is the diffusion, and not the polymer matrix relaxation.

In order to determine the value of diffusion coefficient of Gent, the early time approximation (ETA) model was applied and compared to diffusion coefficient calculated from GFD model. For ETA, two equations were used, standard ETA Eqs. (4) and a modified ETA Eqs. (5) proposed by Ritger and Peppas [27]. According to Ritger and Peppas, the standard ETA is frequently misused, even though it only applies to very specific cases of swelling, and for specific geometries of thin films with very high aspect ratio (diameter divided by thickness; a thin film will typically have aspect ratio of the order of ~100, whereas for thick hydrogel discs it is closer to unity) [27]. In these equations, c_t/c₀ denotes the fraction of released Gent at the time t, D is the diffusion coefficient of Gent during the release, t is the time of release, δ is the hydrogel thickness and r is the radius of the hydrogel disc:

$${\displaystyle \frac{{\mathrm{c}}_{\mathrm{t}}}{{\mathrm{c}}_0}}=4\cdot {\left({\displaystyle \frac{\mathrm{D}\cdot \mathrm{t}}{\mathsf{\pi }\cdot {\mathsf{\delta }}^2}}\right)}^{1/2}$$

(4)

$${\displaystyle \frac{c_t}{c_0}}=4\cdot {\left({\displaystyle \frac{D\cdot t}{\pi \cdot r^2}}\right)}^{1/2}-\pi \cdot \left({\displaystyle \frac{D\cdot t}{\pi \cdot r^2}}\right)-{\displaystyle \frac{\pi }{3}}\cdot {\left({\displaystyle \frac{D\cdot t}{\pi \cdot r^2}}\right)}^{3/2}+4\cdot {\left({\displaystyle \frac{D\cdot t}{\pi \cdot {\delta }^2}}\right)}^{1/2}$$

(5)

$$-{\displaystyle \frac{2r}{\delta }}\cdot \left[8\cdot \left({\displaystyle \frac{D\cdot t}{\pi \cdot r^2}}\right)-2\pi \cdot {\left({\displaystyle \frac{D\cdot t}{\pi \cdot r^2}}\right)}^{3/2}-{\displaystyle \frac{2\pi }{3}}\cdot {\left({\displaystyle \frac{D\cdot t}{\pi \cdot r^2}}\right)}^2\right]$$

(6)

The ETA models represent the dependence of the fraction of released Gent on the square root of the release time while diffusion coefficient (D) of Gent release was determined from the slope of initial linear part of the experimental curve. Using modified ETA model, the values of D was calculated to be 7.16 x10^-8 cm² s^-1 for PVA/Gent hydrogel and 4.29x10^-8 cm² s^-1 for PVA/CHI hydrogel, meaning that the release of Gent is slower from the hydrogel with CHI due to greater number of bonds and consequently more crosslinked polymer matrix [28].

2.3. In Vivo Testing

The PVA/Gent and PVA/CHI/Gent hydrogels have been characterized, based on the material cytotoxicity scale [29], as non-toxic and suitable for in vivo testing [22]. Tests conducted on MTT assays even showed that PVA/CHI/Gent samples promote differentiation and growth of MRC-5 (human fibroblasts), as evidenced by cell viability value higher than 100%. However, in the case of L929 cells, the viability was slightly decreased for all samples, which could be due to the enhanced sensitivity of the mice cell line toward our samples. These results are promising, as the main concern with biomaterials revolves around their immediate application and potential impact on surrounding tissue.

The in vivo analysis aimed to capture all phases of wound repair - inflammation, proliferation, and remodeling - by analyzing the wounds on the 3^rd, 7^th, 14^th, and 21^st day post-injury [30]. Inflammation begins with vascular responses and neutrophil infiltration, later transitioning to macrophage dominance, leading to the proliferative phase, which involves the formation of epithelium and granulation tissue, consisting of fibroblasts, collagen, and new blood vessels [31]. Semi-quantitative scales were used to assess healing parameters, reflecting standard practices in the literature [32,33,34,35].

2.4. Clinical evaluations of the burn wound and wound contraction

Throughout evaluation of PVA/Gent and PVA/CHI/Gent wound healing efficacy on second-degree burns in Wistar rats, the tested hydrogels demonstrated good handling properties and hydrophilicity, allowing easy adherence to the wound bed. During the dressing changes for measurement and photography, the newly formed tissue remained intact, and the hydrogels did not adhere to the wound.

Both hydrogels demonstrated enhanced healing relative to the Ctr group, showing reductions in all clinical parameters (blistering, edema, redness, crust, bleeding, secretion, and scar tissue), notably a significant decrease in redness by day 14^th (p<0.05) (Figure 2 A, B, C, D). Considering that redness (rubor) was defined as one of cardinal signs of inflammation 2000 years ago [36], its reduction confirms that both hydrogels reduce inflammation which is also in accordance with pronounced antibacterial activity of hydrogels containing Gent shown in the 2.1. section.

Wound contraction, a key indicator of healing, was monitored, as an important part of the healing process of full-thickness wounds both in humans and in rats [12]. By the 3^rd day period, the wound area slightly increased in both the Ctr and PVA/Gent groups compared to the initial measurement, while it decreased to a certain extent in the PVA/CHI/Gent treated group potentially due to CHI’s ability to stimulate angiogenesis and fibroblast growth [37,38]. The property of CHI to unable the extension of burn wound in the period of three days post induction is described in literature [39]. Throughout all other time periods, a reduction in wound area compared to the initial measurements was observed in all groups. While no significant differences were observed among groups by the study's end, both PVA/Gent and PVA/CHI/Gent groups consistently exhibited smaller wound areas than the Ctr group, indicating that both hydrogels promote wound repair. This was most pronounced on the 14^th day, with the Ctr group’s wound area measuring 64.47 ± 26.51 mm², compared to 45.55 ± 17.33 mm² in the PVA/Gent group and 47.41 ± 11.03 mm² in the PVA/CHI/Gent treated group (Figure 3).

2.5. Histological Analysis

The histological analysis showed that the effects of PVA/Gent and PVA/CHI/Gent hydrogels on re-epithelization and healing processes in the dermis were not uniform and were significant in different time periods.

Both hydrogels caused improvement of the re-epithelization process, which is depicted by representative microphotographs and graphs in Figure 4. At the 14^th day, a significantly higher percentage of re-epithelization occurred only in the PVA/CHI/Gent treated group compared to the Ctr group (p<0.05) (Figure 4 A, B, C, D). Crust formation was observed in all experimental groups (Figure 4 A, B, C), with no significant differences in the thickness of the newly formed epithelium. In the Ctr group, some animals exhibited all epidermal strata, while others showed only the stratum germinativum, or both the stratum germinativum and stratum granulosum. This pattern was also seen in the PVA/Gent group, however, animals in the PVA/CHI/Gent group showed the presence of all stratums in the newly formed epidermis. During the 21^st day period the higher percentage of re-epithelization was significant in the PVA/Gent treated groups compared to the Ctr group (p<0.05) (Figure 4 E, F, G, H). The crust was present solely in the Ctr group (Figure 4 E), and as in previous periods, there were no significant differences in the thickness of the newly formed epithelium among the experimental groups. By the 21^st day, every animal in each group exhibited all stratums in the newly formed epidermis.

Various studies examining the effectiveness of hydrogels containing different combinations of PVA, CHI and Gent with or without other polymers showed beneficial effect on the process of re-epithelization [15,40,41]. To the best of our knowledge this study is the first explaining the beneficial effects of the exact combination of PVA/Gent and PVA/CHI/Gent hydrogels. The enhanced re-epithelization could be explained by the fact that moist environment enables the faster migration of keratinocytes [42]. The effects were evident earlier in the PVA/CHI/Gent treated group which can be explained by the synergy between the moist environment and CHI activating various cells (neutrophils, macrophages and fibroblasts) to produce cytokines which enables optimal condition for keratinocytes migration [43].

The individual scores for all parameters assessed in the dermis of the burn wound (subepithelial neutrophils, fibroblasts, collagen deposition, edema and angiogenesis) are graphically represented in heat maps in Figure 5, while the representative photomicrographs of the assessments are presented in Figure 6.

At the 3rd day period, all parameters in the treated groups showed higher grades, except for edema, which was rated higher in the Ctr group (Figure 5 A, B, C). Significant differences were noted for fibroblast count and collagen deposition, with the PVA/Gent treated group showing notably higher levels compared to the Ctr group (p<0.05) (Figure 5 A, B, C and Figure 6 B3, B4, F3, F4). These findings suggest that the PVA/Gent hydrogel positively influences early wound healing by enhancing fibroblast proliferation and collagen synthesis. This observation aligns with in vitro studies where PVA hydrogels, even without added growth factors, induced slight increases in fibroblast proliferation and facilitated transit of cells [44]. The more pronounced increase in fibroblasts observed in our study may be attributed to the in vivo setting and the presence of different cells, namely neutrophils during the inflammatory phase. Neutrophils protect the wound from infection and clear tissue debris [45], while also stimulating fibroblast proliferation both directly, via cytokine expression (IL-8, IL-1β, MCP-1), and indirectly by attracting macrophages, which further enhance fibroblast activity [46]. The significant increase in collagen deposition in the PVA/Gent group on the 3^rd day is likely a direct consequence of the elevated fibroblast presence, considering the fact that the collagen is mostly synthesized by fibroblasts [45].

By the 7^th day, edema remained more prominent in the Ctr group, which also exhibited higher grades of subepithelial neutrophils (Figure 5 D, E, F). Although fibroblast count, collagen deposition, and angiogenesis continued to be more pronounced in the treated groups, no significant differences were observed at this time point. This lack of significance may reflect the transitional phase of wound healing, where the inflammatory response begins to subside, and proliferative processes take precedence.

The trend persisted into the 14^th day period, where the Ctr group had higher grades of subepithelial neutrophils and edema, with significant reductions in these parameters in the PVA/CHI/Gent group (p<0.05) (Figure 5 G, H, I and Figure 6 D2, D4, L2, L4). Fibroblast count, collagen deposition, and angiogenesis remained more pronounced in the treated groups, with collagen deposition significantly higher in the PVA/CHI/Gent group (p<0.05) (Figure 5 H, I, J and Figure 6 D4, L4). The delayed, but significant effects of the PVA/CHI/Gent hydrogel likely result from the synergistic antimicrobial action of CHI and Gent. Chitosan’s antimicrobial activity, based on its positive charge interacting with negatively charged bacterial cell membranes, enhances permeability, while Gent’s limited cell penetration is overcome by CHI’s effect [47,48]. Reduced subepithelial neutrophil presence and edema in the PVA/CHI/Gent group may thus reflect decreased inflammation due to the combined antibacterial action, which is consistent with our antibacterial activity tests and supported by literature [49]. Such enhanced antibacterial effects are desirable in wound dressings to prevent biofilm formation and bacterial adhesion during the critical early stages of healing. Also, the time-dependent release of the antibacterial agent would ensure prolonged sterility of the dressing and the wound itself. Moreover, the significant increase in collagen deposition at 14^th day period in the PVA/CHI/Gent group aligns with previous findings where CHI stimulated collagen synthesis by enhancing prolyl hydroxylase activity in granulation tissue [50]. This enzyme’s peak activity at the 14^th post-implantation day matches the elevated collagen levels observed in our study, reinforcing the efficacy of the PVA/CHI/Gent hydrogel in supporting the proliferative phase of wound healing.

By the 21^st day, edema was no longer visible in any observed group (Figure 5 J, K, L and Figure 6 E4, I4, M4). Although the Ctr group still exhibited more subepithelial neutrophils in the dermis, collagen deposition was rated higher in this group for the first time (Figure 5 J, K, L). Fibroblast presence and angiogenesis consistently remained higher in the treated groups, though none of the differences during this period reached significance. These results suggest that while initial inflammation is mitigated by the hydrogel treatments, long-term matrix remodeling processes like collagen deposition may vary depending on the wound environment and treatment dynamics.

3. Conclusions

The novel PVA/Gent and PVA/CHI/Gent hydrogels demonstrated significant wound healing efficacy in second-degree burn wounds. The hydrogels showed good handling and hydrophilicity adhering well to the wound bed. Both hydrogels had significantly reduced inflammation and exhibited strong antibacterial properties. They enhanced re-epithelization and increased fibroblast proliferation and collagen deposition in the healing process. Both PVA/Gent and PVA/CHI/Gent hydrogels show potential as effective wound dressings, promoting faster healing and improved tissue recovery.

4. Materials and Methods

4.1. Materials

The following chemicals were utilized for preparation of PVA/Gent and PVA/CHI/Gent hydrogels: poly(vinyl alcohol) powder (fully hydrolyzed, Mw = 70-100 kDa, Sigma Aldrich, USA), chitosan powder (Mw = 190-310 kDa, deacetylation degree 75-85 %, Sigma Aldrich, USA) and gentamicin sulfate solution (50 mg/ml in dH₂O, Sigma Aldrich, USA). Deionized water was obtained by passing the distilled water through a GenPure ultrapure water system (TKA, Germany).

4.1.1. Synthesis of PVA/Gent Hydrogel

Colloid dispersion of PVA (10 wt.%) was prepared by dissolving PVA powder in hot distilled water at 90ºC for 2h, under magnetic stirring. The PVA hydrogels were obtained by physical cross linking of PVA dispersion using freezing-thawing method in 5 cycles. One cycle consisted of freezing for 16h at -18ºC, followed by thawing for 8h at 4ºC. Thus, obtained hydrogels were cut into discs with diameters, d, of 10 mm and thicknesses, δ, of 4 mm. Then, the hydrogels were swollen in 5.0 mg/ml gentamicin solution at 37^oC during 48h, to obtain PVA/Gent hydrogels.

4.1.2. Synthesis of PVA/CHI/Gent Hydrogel

Colloid dispersion of PVA was prepared by dissolving PVA powder in hot distilled water at 90ºC for 2h, under magnetic stirring. Chitosan was dissolved in 2 vol% CH₃COOH under constant stirring at room temperature. After cooling of PVA, the CHI dispersion was added dropwise and the final dispersions (containing 10 wt.% PVA and 0.5 wt.% CHI) were homogenized by mixing at room temperature for 2-3h. Further, the PVA/CHI hydrogels were prepared by physical cross linking of PVA/CHI dispersion using freezing-thawing method in 5 cycles. One cycle consisted of 16h freezing at -18°C followed by 8h thawing at 4°C. Finally, the hydrogels were swollen in 5.0 mg/ml gentamicin solution at 37^oC during 48h, to obtain PVA/CHI/Gent hydrogels.

4.1.3. Antibacterial Activity

Antibacterial activity of PVA/Gent and PVA/CHI/Gent hydrogels was investigated against Gram-negative and Gram-positive bacteria strains – Escherichia coli ATCC 25922 and Staphylococcus aureus TL (culture collection Faculty of Technology and Metallurgy, University of Belgrade, Serbia). Hydrogels were sterilized in the laminar air flow chamber by exposure to a UV-C lamp (30 min, 60 cm lamp distance) and their antibacterial activity was evaluated by disc-diffusion method. A soft-top agar (0.7 wt.%, 15 ml) was melted, cooled down to ∼55°C and inoculated with bacteria cultures (150 μl) overnight (∼18h), then gently stirred and poured over previously solidified nutrient agar base in sterile Petri dishes. Number of bacteria in the nutrient soft-top agar layer was set to be ∼10⁶ CFU ml⁻¹. After solidification of soft-top agar, hydrogel disc samples were placed on its surface. The inhibition zone widths were measured after 24h incubation at 37°C.

4.1.4. Gentamicin Release

High-performance liquid chromatography (HPLC) (Thermo Fisher Scientific, USA) was utilized for Gent components separation and the detection and quantitative analysis were done in an ion trap mass spectrometer (MS) (LCQ Advantage, Thermo Fisher Scientific). High-performance liquid chromatography was equipped with a reverse-phase column (4.6 mm × 75 mm × 3.5 μm) Zorbax Eclipse® XDB-C18 (Agilent Technologies, USA), in front of which a precolumn (4.6 mm × 12.5 mm × 5 μm) was placed. Methanol (A), deionized water (B), and 10% acetic acid (C) comprised the mobile phase. The optimized HPLC and MS operating parameters (mobile-phase gradient, analytes’ precursor ions, fragmentation reactions used for quantification, and optimal collision energies) for the determination of Gent compounds were published in our previous paper [51]. The gentamicin mass spectra were collected in the m/z range of 50-1000. As expected, the MS spectrum revealed the three most abundant ions since Gent is composed of three compounds – Gent C1a, C2, and C1. These ions were further chosen as the precursor ions for each compound. Their most sensitive transitions were selected for quantification purposes. The presented Gent concentrations represent sums of the three determined Gent compounds.

4.2. Animals

Three-month-old male Wistar rats weighing 300-330g obtained from the Department of Laboratory and Experimental Care and Use of Animals Unit of the Institute of Medical Research, Military Medical Academy (Belgrade, Serbia) were used in the experiment. Rats were housed individually in polypropylene cages in an air-conditioned animal facility, under standard conditions with a photoperiodic cycle of 12h light: 12h darkness, and food and water intake ad libitum. All of the experimental procedures were approved by the Ethical Committee of the Faculty of Veterinary Medicine University of Belgrade and by the Ministry of agriculture, forestry and water management – veterinary administration (decision number 323-07-04903/2022-05/1). The experiment and animal handling were carried out in accordance with the ARRIVE guidelines and the European Union’s Directive 2010/63/EU on the Protection of Animals Used for Scientific Purposes.

4.2.1. Experimental Design

Burn wounds were modeled following the protocol by Tavares Pereira et al. [30]. The procedure utilized general anesthesia with intraperitoneal injections of 75 mg/kg ketamine hydrochloride (Ketamidor 10%, 100mg/ml, RICHTER PHARMA AG, Austria) and 10 mg/kg xylazine (Xylased 2%, BIOVETA, Czech Republic). After trichotomy of the back skin and disinfection with povidone-iodine, a thermal injury was induced by applying a solid aluminum bar (10 mm diameter, mass of 51g), previously heated in boiling water (temperature of 100°C), to the skin of the dorsal proximal region for 15 seconds without any additional pressure. Post-procedure, animals received intramuscular analgesia with 5 mg/kg Ketoprofen (Ketonal, á 100mg/2ml, SANDOZ, Switzerland), for three consecutive days.

Following the burns, animals were randomly assigned to three experimental groups: control (Ctr) (n=16), PVA/Gent treated (n=16), and PVA/CHI/Gent treated group (n=16), with daily dressing of the wounds. Clinical evaluations of the burn wounds were conducted on the 3^rd, 7^th, 14^th, and 21^st day by semi-quantitatively assessing the parameters such as blistering, edema, redness, crust, bleeding, secretion, and scar tissue, on a scale from 0 (absent) to 3 (severe). Diameters of the burn were measured immediately after burn induction and on previously specified days to calculate wound contraction using the Eqs. (6):

$$X_{\%}=\left[X_0-X/X_0\right]\ast 100$$

(2)

Here, X_% is percentage of wound contraction on the specified day, X₀ is the initial diameter of the wound and X is diameter of the wound on the specified day. On those evaluation days, four animals from each group were euthanized using 100 mg/kg Euthasol Euthanasia Solution (Produlab Pharma Production B.V., Raamsdonksveer, Netherlands) for histological sample collection.

4.2.2. Histological Analysis

During sampling, a square piece of skin containing the entire burn wound was taken and bisected along the diameter of the circular burn. Both halves of each sample were fixed and processed by conventional manner for histological examination. Paraffin-embedded tissue blocks were serially sectioned and stained with Hematoxylin-Eosin (H/E) and Masson Goldner (MG) staining kit (Merck Millipore, Darmstadt, Germany), then evaluated on a standard Olympus CX31 microscope equipped with a digital camera and software (UC50 Soft Imaging Solutions camera and SensEntry 1.13 software, Münster, Germany). Histological analyses included the assessments of re-epithelization at the 14^th and 21^st day and evaluations of the dermis at all time points.

Re-epithelization process was evaluated on H/E stained sections at 4x magnification by calculating the percentage of re-epithelization X_e% on the specified day using the formula: $X_{e\%}=\left(X_e/X_e+X\right)\ast 100$ , where X_e is the length of newly formed epithelium on the specified day and X is the length of open wound on the specified day. Additionally, the thickness of the newly formed epithelium was measured at five points; approximately 300 µm from the wound edges on both sides of the burn wound, at 40x magnification and the stratification in the epithelium was assessed.

The semi-quantitative assessment of the dermis evaluated the percentage of dermal coverage by subepithelial neutrophils, fibroblasts, collagen deposition, edema, and angiogenesis at 40x magnification. Each parameter was scored as absent (0) for up to 10%, mild (1) for 10-40%, moderate (2) for 40-70%, and severe (3) for 70-100% coverage. All parameters were assessed on H/E stained sections, except collagen deposition, which was estimated on MG stained sections.

4.3. Statistical Analysis

All collected data were analyzed in the GraphPad Prism 9 software (GraphPad, San Diego, CA, USA). Prior to the applying of the statistical analysis the normality of data distribution was tested using the Shapiro-Wilk normality test. All histological examination results were analyzed using the Kruskal-Wallis test followed by Dunn’s post-hoc test, except for the re-epithelization assessment on day 14^th, which was analyzed using one-way ANOVA followed by Tukey’s post-hoc test.